[rrd-users] intermittent graphing wrong RRA

[Kathy Kessey]

I ran into this strange problem that I can't seem to pin down, wondering
if anybody has any ideas.

I have 2 different RRD formats:

ATM-PVC:
rrdtool create [fn] --start [st] --step 900 [ 7 DS's heartbeat = 2700 ]
RRA:AVERAGE:0.5:1:768
RRA:AVERAGE:0.5:8:768
RRA:AVERAGE:0.5:96:400
[ plus identical MAX RRA's ]

ATM-PORT:
rrdtool create [fn] --start [st] --step 900 [ 74 DS's heartbeat = 2700 ]
RRA:AVERAGE:0.5:1:768
RRA:AVERAGE:0.5:8:768
RRA:AVERAGE:0.5:96:400
[ plus identical MAX RRA's ]

ok, so my RRAs are 8 days of 15min samples, then averaged to 2hr samples
for 64 days, then averaged to 24hr for 400 days.  The only config
difference between the 2 is the large number of DS in the port RRDs (7 vs
74.)

I've had the pvc data collecting for a couple of months, works great.
I just started collecting the port data, and after the first 8 days,
occasionally a one day (-1d) graph for any DS shows 2hr samples.  Then
later, its back to showing 15min samples.  When the graph is showing  2hr
samples, I've fetched the data (-1d) and every sample returned is a
correct 15min interval with unique data in each sample.

I've used RRD a lot, for a variety of different tools, and never been able
to not find my mistake, so I've looked for the usual stupid type-o's, etc.
Any suggestions would be greatly appreciated!

thanks,

Kathy




Kathy Kessey
New Edge Networks
360.906.9901




--
Unsubscribe mailto:rrd-users-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-users-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi